[apple/xnu.git] / osfmk / kern / lock.c

/*
 * Copyright (c) 2000 Apple Computer, Inc. All rights reserved.
 *
 * @APPLE_LICENSE_HEADER_START@
 * 
 * The contents of this file constitute Original Code as defined in and
 * are subject to the Apple Public Source License Version 1.1 (the
 * "License").  You may not use this file except in compliance with the
 * License.  Please obtain a copy of the License at
 * http://www.apple.com/publicsource and read it before using this file.
 * 
 * This Original Code and all software distributed under the License are
 * distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY KIND, EITHER
 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT.  Please see the
 * License for the specific language governing rights and limitations
 * under the License.
 * 
 * @APPLE_LICENSE_HEADER_END@
 */
/*
 * @OSF_COPYRIGHT@
 */
/* 
 * Mach Operating System
 * Copyright (c) 1991,1990,1989,1988,1987 Carnegie Mellon University
 * All Rights Reserved.
 * 
 * Permission to use, copy, modify and distribute this software and its
 * documentation is hereby granted, provided that both the copyright
 * notice and this permission notice appear in all copies of the
 * software, derivative works or modified versions, and any portions
 * thereof, and that both notices appear in supporting documentation.
 * 
 * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS"
 * CONDITION.  CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND FOR
 * ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE.
 * 
 * Carnegie Mellon requests users of this software to return to
 * 
 *  Software Distribution Coordinator  or  Software.Distribution@CS.CMU.EDU
 *  School of Computer Science
 *  Carnegie Mellon University
 *  Pittsburgh PA 15213-3890
 * 
 * any improvements or extensions that they make and grant Carnegie Mellon
 * the rights to redistribute these changes.
 */
/*
 *      File:   kern/lock.c
 *      Author: Avadis Tevanian, Jr., Michael Wayne Young
 *      Date:   1985
 *
 *      Locking primitives implementation
 */

#include <cpus.h>
#include <mach_kdb.h>
#include <mach_ldebug.h>

#include <kern/lock.h>
#include <kern/etap_macros.h>
#include <kern/misc_protos.h>
#include <kern/thread.h>
#include <kern/sched_prim.h>
#include <kern/xpr.h>
#include <kern/debug.h>
#include <string.h>

#if     MACH_KDB
#include <ddb/db_command.h>
#include <ddb/db_output.h>
#include <ddb/db_sym.h>
#include <ddb/db_print.h>
#endif  /* MACH_KDB */

#ifdef __ppc__
#include <ppc/Firmware.h>
#include <ppc/POWERMAC/mp/MPPlugIn.h>
#endif

#define ANY_LOCK_DEBUG  (USLOCK_DEBUG || LOCK_DEBUG || MUTEX_DEBUG)

/*
 *      Some portions of the lock debugging code must run with
 *      interrupts disabled.  This can be machine-dependent,
 *      but we don't have any good hooks for that at the moment.
 *      If your architecture is different, add a machine-dependent
 *      ifdef here for these macros.            XXX
 */

#define DISABLE_INTERRUPTS(s)   s = ml_set_interrupts_enabled(FALSE)
#define ENABLE_INTERRUPTS(s)    (void)ml_set_interrupts_enabled(s)

#if     NCPUS > 1
/* Time we loop without holding the interlock. 
 * The former is for when we cannot sleep, the latter
 * for when our thread can go to sleep (loop less)
 * we shouldn't retake the interlock at all frequently
 * if we cannot go to sleep, since it interferes with
 * any other processors. In particular, 100 is too small
 * a number for powerpc MP systems because of cache
 * coherency issues and differing lock fetch times between
 * the processors
 */
unsigned int lock_wait_time[2] = { (unsigned int)-1, 100 } ;
#else   /* NCPUS > 1 */

        /*
         *      It is silly to spin on a uni-processor as if we
         *      thought something magical would happen to the
         *      want_write bit while we are executing.
         */

unsigned int lock_wait_time[2] = { 0, 0 };
#endif  /* NCPUS > 1 */

/* Forwards */

#if     MACH_KDB
void    db_print_simple_lock(
                        simple_lock_t   addr);

void    db_print_mutex(
                        mutex_t         * addr);
#endif  /* MACH_KDB */


#if     USLOCK_DEBUG
/*
 *      Perform simple lock checks.
 */
int     uslock_check = 1;
int     max_lock_loops  = 100000000;
decl_simple_lock_data(extern , printf_lock)
decl_simple_lock_data(extern , panic_lock)
#if     MACH_KDB && NCPUS > 1
decl_simple_lock_data(extern , kdb_lock)
#endif  /* MACH_KDB && NCPUS >1 */
#endif  /* USLOCK_DEBUG */


/*
 *      We often want to know the addresses of the callers
 *      of the various lock routines.  However, this information
 *      is only used for debugging and statistics.
 */
typedef void    *pc_t;
#define INVALID_PC      ((void *) VM_MAX_KERNEL_ADDRESS)
#define INVALID_THREAD  ((void *) VM_MAX_KERNEL_ADDRESS)
#if     ANY_LOCK_DEBUG || ETAP_LOCK_TRACE
#define OBTAIN_PC(pc,l) ((pc) = (void *) GET_RETURN_PC(&(l)))
#else   /* ANY_LOCK_DEBUG || ETAP_LOCK_TRACE */
#ifdef  lint
/*
 *      Eliminate lint complaints about unused local pc variables.
 */
#define OBTAIN_PC(pc,l) ++pc
#else   /* lint */
#define OBTAIN_PC(pc,l)
#endif  /* lint */
#endif  /* USLOCK_DEBUG || ETAP_LOCK_TRACE */


/* #ifndef      USIMPLE_LOCK_CALLS
 * The i386 production version of usimple_locks isn't ready yet.
 */
/*
 *      Portable lock package implementation of usimple_locks.
 */

#if     ETAP_LOCK_TRACE
#define ETAPCALL(stmt)  stmt
void            etap_simplelock_init(simple_lock_t, etap_event_t);
void            etap_simplelock_unlock(simple_lock_t);
void            etap_simplelock_hold(simple_lock_t, pc_t, etap_time_t);
etap_time_t     etap_simplelock_miss(simple_lock_t);

void            etap_mutex_init(mutex_t*, etap_event_t);
void            etap_mutex_unlock(mutex_t*);
void            etap_mutex_hold(mutex_t*, pc_t, etap_time_t);
etap_time_t     etap_mutex_miss(mutex_t*);
#else   /* ETAP_LOCK_TRACE */
#define ETAPCALL(stmt)
#endif  /* ETAP_LOCK_TRACE */

#if     USLOCK_DEBUG
#define USLDBG(stmt)    stmt
void            usld_lock_init(usimple_lock_t, etap_event_t);
void            usld_lock_pre(usimple_lock_t, pc_t);
void            usld_lock_post(usimple_lock_t, pc_t);
void            usld_unlock(usimple_lock_t, pc_t);
void            usld_lock_try_pre(usimple_lock_t, pc_t);
void            usld_lock_try_post(usimple_lock_t, pc_t);
void            usld_lock_held(usimple_lock_t);
void            usld_lock_none_held(void);
int             usld_lock_common_checks(usimple_lock_t, char *);
#else   /* USLOCK_DEBUG */
#define USLDBG(stmt)
#endif  /* USLOCK_DEBUG */

/*
 *      Initialize a usimple_lock.
 *
 *      No change in preemption state.
 */
void
usimple_lock_init(
        usimple_lock_t  l,
        etap_event_t    event)
{
        USLDBG(usld_lock_init(l, event));
        ETAPCALL(etap_simplelock_init((l),(event)));
        hw_lock_init(&l->interlock);
}


/*
 *      Acquire a usimple_lock.
 *
 *      Returns with preemption disabled.  Note
 *      that the hw_lock routines are responsible for
 *      maintaining preemption state.
 */
void
usimple_lock(
        usimple_lock_t  l)
{
        int i;
        pc_t            pc;
#if     ETAP_LOCK_TRACE
        etap_time_t     start_wait_time;
        int             no_miss_info = 0;
#endif  /* ETAP_LOCK_TRACE */
#if     USLOCK_DEBUG
        int             count = 0;
#endif  /* USLOCK_DEBUG */

        OBTAIN_PC(pc, l);
        USLDBG(usld_lock_pre(l, pc));
#if     ETAP_LOCK_TRACE
        ETAP_TIME_CLEAR(start_wait_time);
#endif  /* ETAP_LOCK_TRACE */

#ifdef __ppc__
        if(!hw_lock_to(&l->interlock, LockTimeOut)) {   /* Try to get the lock with a timeout */ 
        
                panic("simple lock deadlock detection - l=%08X, cpu=%d, ret=%08X", l, cpu_number(), pc);

#else /* __ppc__ */
        while (!hw_lock_try(&l->interlock)) {
                ETAPCALL(if (no_miss_info++ == 0)
                        start_wait_time = etap_simplelock_miss(l));
                while (hw_lock_held(&l->interlock)) {
                        /*
                         *      Spin watching the lock value in cache,
                         *      without consuming external bus cycles.
                         *      On most SMP architectures, the atomic
                         *      instruction(s) used by hw_lock_try
                         *      cost much, much more than an ordinary
                         *      memory read.
                         */
#if     USLOCK_DEBUG
                        if (count++ > max_lock_loops
#if     MACH_KDB && NCPUS > 1
                            && l != &kdb_lock
#endif  /* MACH_KDB && NCPUS > 1 */
                            ) {
                                if (l == &printf_lock) {
                                        return;
                                }
                                mp_disable_preemption();
                                panic("simple lock deadlock detection - l=%08X (=%08X), cpu=%d, ret=%08X", 
                                   l, *hw_lock_addr(l->interlock), cpu_number(), pc);
                                count = 0;
                                mp_enable_preemption();
                        }
#endif  /* USLOCK_DEBUG */
                }
#endif /* 0 */
        }
        ETAPCALL(etap_simplelock_hold(l, pc, start_wait_time));
        USLDBG(usld_lock_post(l, pc));
}


/*
 *      Release a usimple_lock.
 *
 *      Returns with preemption enabled.  Note
 *      that the hw_lock routines are responsible for
 *      maintaining preemption state.
 */
void
usimple_unlock(
        usimple_lock_t  l)
{
        pc_t    pc;

//      checkNMI();                                                                             /* (TEST/DEBUG) */
                
        OBTAIN_PC(pc, l);
        USLDBG(usld_unlock(l, pc));
        ETAPCALL(etap_simplelock_unlock(l));
        hw_lock_unlock(&l->interlock);
}


/*
 *      Conditionally acquire a usimple_lock.
 *
 *      On success, returns with preemption disabled.
 *      On failure, returns with preemption in the same state
 *      as when first invoked.  Note that the hw_lock routines
 *      are responsible for maintaining preemption state.
 *
 *      XXX No stats are gathered on a miss; I preserved this
 *      behavior from the original assembly-language code, but
 *      doesn't it make sense to log misses?  XXX
 */
unsigned int
usimple_lock_try(
        usimple_lock_t  l)
{
        pc_t            pc;
        unsigned int    success;
        etap_time_t     zero_time;

        OBTAIN_PC(pc, l);
        USLDBG(usld_lock_try_pre(l, pc));
        if (success = hw_lock_try(&l->interlock)) {
                USLDBG(usld_lock_try_post(l, pc));
                ETAP_TIME_CLEAR(zero_time);
                ETAPCALL(etap_simplelock_hold(l, pc, zero_time));
        }
        return success;
}

#if ETAP_LOCK_TRACE
void
simple_lock_no_trace(
        simple_lock_t   l)
{
        pc_t            pc;

        OBTAIN_PC(pc, l);
        USLDBG(usld_lock_pre(l, pc));
        while (!hw_lock_try(&l->interlock)) {
                while (hw_lock_held(&l->interlock)) {
                        /*
                         *      Spin watching the lock value in cache,
                         *      without consuming external bus cycles.
                         *      On most SMP architectures, the atomic
                         *      instruction(s) used by hw_lock_try
                         *      cost much, much more than an ordinary
                         *      memory read.
                         */
                }
        }
        USLDBG(usld_lock_post(l, pc));
}

void
simple_unlock_no_trace(
        simple_lock_t   l)
{
        pc_t    pc;

        OBTAIN_PC(pc, l);
        USLDBG(usld_unlock(l, pc));
        hw_lock_unlock(&l->interlock);
}

int
simple_lock_try_no_trace(
        simple_lock_t   l)
{
        pc_t            pc;
        unsigned int    success;

        OBTAIN_PC(pc, l);
        USLDBG(usld_lock_try_pre(l, pc));
        if (success = hw_lock_try(&l->interlock)) {
                USLDBG(usld_lock_try_post(l, pc));
        }
        return success;
}
#endif  /* ETAP_LOCK_TRACE */


#if     USLOCK_DEBUG
/*
 *      Verify that the lock is locked and owned by
 *      the current thread.
 */
void
usimple_lock_held(
        usimple_lock_t  l)
{
        usld_lock_held(l);
}


/*
 *      Verify that no usimple_locks are held by
 *      this processor.  Typically used in a
 *      trap handler when returning to user mode
 *      or in a path known to relinquish the processor.
 */
void
usimple_lock_none_held(void)
{
        usld_lock_none_held();
}
#endif  /* USLOCK_DEBUG */


#if     USLOCK_DEBUG
/*
 *      States of a usimple_lock.  The default when initializing
 *      a usimple_lock is setting it up for debug checking.
 */
#define USLOCK_CHECKED          0x0001          /* lock is being checked */
#define USLOCK_TAKEN            0x0002          /* lock has been taken */
#define USLOCK_INIT             0xBAA0          /* lock has been initialized */
#define USLOCK_INITIALIZED      (USLOCK_INIT|USLOCK_CHECKED)
#define USLOCK_CHECKING(l)      (uslock_check &&                        \
                                 ((l)->debug.state & USLOCK_CHECKED))

/*
 *      Maintain a per-cpu stack of acquired usimple_locks.
 */
void    usl_stack_push(usimple_lock_t, int);
void    usl_stack_pop(usimple_lock_t, int);

/*
 *      Trace activities of a particularly interesting lock.
 */
void    usl_trace(usimple_lock_t, int, pc_t, const char *);


/*
 *      Initialize the debugging information contained
 *      in a usimple_lock.
 */
void
usld_lock_init(
        usimple_lock_t  l,
        etap_event_t    type)
{
        if (l == USIMPLE_LOCK_NULL)
                panic("lock initialization:  null lock pointer");
        l->lock_type = USLOCK_TAG;
        l->debug.state = uslock_check ? USLOCK_INITIALIZED : 0;
        l->debug.lock_cpu = l->debug.unlock_cpu = 0;
        l->debug.lock_pc = l->debug.unlock_pc = INVALID_PC;
        l->debug.lock_thread = l->debug.unlock_thread = INVALID_THREAD;
        l->debug.duration[0] = l->debug.duration[1] = 0;
        l->debug.unlock_cpu = l->debug.unlock_cpu = 0;
        l->debug.unlock_pc = l->debug.unlock_pc = INVALID_PC;
        l->debug.unlock_thread = l->debug.unlock_thread = INVALID_THREAD;
}


/*
 *      These checks apply to all usimple_locks, not just
 *      those with USLOCK_CHECKED turned on.
 */
int
usld_lock_common_checks(
        usimple_lock_t  l,
        char            *caller)
{
        if (l == USIMPLE_LOCK_NULL)
                panic("%s:  null lock pointer", caller);
        if (l->lock_type != USLOCK_TAG)
                panic("%s:  0x%x is not a usimple lock", caller, (integer_t) l);
        if (!(l->debug.state & USLOCK_INIT))
                panic("%s:  0x%x is not an initialized lock",
                      caller, (integer_t) l);
        return USLOCK_CHECKING(l);
}


/*
 *      Debug checks on a usimple_lock just before attempting
 *      to acquire it.
 */
/* ARGSUSED */
void
usld_lock_pre(
        usimple_lock_t  l,
        pc_t            pc)
{
        char            *caller = "usimple_lock";


#if 0
        printf("*** %08X %08X %04X %02X %08X %02X %08X - %s\n", /* (TEST/DEBUG) */
                l->debug.lock_pc,
                l->debug.lock_thread,
                l->debug.state,
                l->debug.lock_cpu,
                l->debug.unlock_thread,
                l->debug.unlock_cpu,
                l->debug.unlock_pc,
                caller);
#endif

        if (!usld_lock_common_checks(l, caller))
                return;

/*
 *      Note that we have a weird case where we are getting a lock when we are]
 *      in the process of putting the system to sleep. We are running with no
 *      current threads, therefore we can't tell if we are trying to retake a lock
 *      we have or someone on the other processor has it.  Therefore we just
 *      ignore this test if the locking thread is 0.
 */

        if ((l->debug.state & USLOCK_TAKEN) && l->debug.lock_thread &&
            l->debug.lock_thread == (void *) current_thread()) {
                printf("%s:  lock 0x%x already locked (at 0x%x) by",
                      caller, (integer_t) l, l->debug.lock_pc);
                printf(" current thread 0x%x (new attempt at pc 0x%x)\n",
                       l->debug.lock_thread, pc);
                panic(caller);
        }
        mp_disable_preemption();
        usl_trace(l, cpu_number(), pc, caller);
        mp_enable_preemption();
}


/*
 *      Debug checks on a usimple_lock just after acquiring it.
 *
 *      Pre-emption has been disabled at this point,
 *      so we are safe in using cpu_number.
 */
void
usld_lock_post(
        usimple_lock_t  l,
        pc_t            pc)
{
        register int    mycpu;
        char            *caller = "successful usimple_lock";


#if 0
        printf("*** %08X %08X %04X %02X %08X %02X %08X - %s\n", /* (TEST/DEBUG) */
                l->debug.lock_pc,
                l->debug.lock_thread,
                l->debug.state,
                l->debug.lock_cpu,
                l->debug.unlock_thread,
                l->debug.unlock_cpu,
                l->debug.unlock_pc,
                caller);
#endif

        if (!usld_lock_common_checks(l, caller))
                return;

        if (!((l->debug.state & ~USLOCK_TAKEN) == USLOCK_INITIALIZED))
                panic("%s:  lock 0x%x became uninitialized",
                      caller, (integer_t) l);
        if ((l->debug.state & USLOCK_TAKEN))
                panic("%s:  lock 0x%x became TAKEN by someone else",
                      caller, (integer_t) l);

        mycpu = cpu_number();
        l->debug.lock_thread = (void *)current_thread();
        l->debug.state |= USLOCK_TAKEN;
        l->debug.lock_pc = pc;
        l->debug.lock_cpu = mycpu;

        usl_stack_push(l, mycpu);
        usl_trace(l, mycpu, pc, caller);
}


/*
 *      Debug checks on a usimple_lock just before
 *      releasing it.  Note that the caller has not
 *      yet released the hardware lock.
 *
 *      Preemption is still disabled, so there's
 *      no problem using cpu_number.
 */
void
usld_unlock(
        usimple_lock_t  l,
        pc_t            pc)
{
        register int    mycpu;
        char            *caller = "usimple_unlock";


#if 0
        printf("*** %08X %08X %04X %02X %08X %02X %08X - %s\n", /* (TEST/DEBUG) */
                l->debug.lock_pc,
                l->debug.lock_thread,
                l->debug.state,
                l->debug.lock_cpu,
                l->debug.unlock_thread,
                l->debug.unlock_cpu,
                l->debug.unlock_pc,
                caller);
#endif

        if (!usld_lock_common_checks(l, caller))
                return;

        mycpu = cpu_number();

        if (!(l->debug.state & USLOCK_TAKEN))
                panic("%s:  lock 0x%x hasn't been taken",
                      caller, (integer_t) l);
        if (l->debug.lock_thread != (void *) current_thread())
                panic("%s:  unlocking lock 0x%x, owned by thread 0x%x",
                      caller, (integer_t) l, l->debug.lock_thread);
        if (l->debug.lock_cpu != mycpu) {
                printf("%s:  unlocking lock 0x%x on cpu 0x%x",
                       caller, (integer_t) l, mycpu);
                printf(" (acquired on cpu 0x%x)\n", l->debug.lock_cpu);
                panic(caller);
        }
        usl_trace(l, mycpu, pc, caller);
        usl_stack_pop(l, mycpu);

        l->debug.unlock_thread = l->debug.lock_thread;
        l->debug.lock_thread = INVALID_PC;
        l->debug.state &= ~USLOCK_TAKEN;
        l->debug.unlock_pc = pc;
        l->debug.unlock_cpu = mycpu;
}


/*
 *      Debug checks on a usimple_lock just before
 *      attempting to acquire it.
 *
 *      Preemption isn't guaranteed to be disabled.
 */
void
usld_lock_try_pre(
        usimple_lock_t  l,
        pc_t            pc)
{
        char            *caller = "usimple_lock_try";

        if (!usld_lock_common_checks(l, caller))
                return;
        mp_disable_preemption();
        usl_trace(l, cpu_number(), pc, caller);
        mp_enable_preemption();
}


/*
 *      Debug checks on a usimple_lock just after
 *      successfully attempting to acquire it.
 *
 *      Preemption has been disabled by the
 *      lock acquisition attempt, so it's safe
 *      to use cpu_number.
 */
void
usld_lock_try_post(
        usimple_lock_t  l,
        pc_t            pc)
{
        register int    mycpu;
        char            *caller = "successful usimple_lock_try";

        if (!usld_lock_common_checks(l, caller))
                return;

        if (!((l->debug.state & ~USLOCK_TAKEN) == USLOCK_INITIALIZED))
                panic("%s:  lock 0x%x became uninitialized",
                      caller, (integer_t) l);
        if ((l->debug.state & USLOCK_TAKEN))
                panic("%s:  lock 0x%x became TAKEN by someone else",
                      caller, (integer_t) l);

        mycpu = cpu_number();
        l->debug.lock_thread = (void *) current_thread();
        l->debug.state |= USLOCK_TAKEN;
        l->debug.lock_pc = pc;
        l->debug.lock_cpu = mycpu;

#if 0
        printf("*** %08X %08X %04X %02X %08X %02X %08X - %s\n", /* (TEST/DEBUG) */
                l->debug.lock_pc,
                l->debug.lock_thread,
                l->debug.state,
                l->debug.lock_cpu,
                l->debug.unlock_thread,
                l->debug.unlock_cpu,
                l->debug.unlock_pc,
                caller);
#endif

        usl_stack_push(l, mycpu);
        usl_trace(l, mycpu, pc, caller);
}


/*
 *      Determine whether the lock in question is owned
 *      by the current thread.
 */
void
usld_lock_held(
        usimple_lock_t  l)
{
        char            *caller = "usimple_lock_held";


#if 0
        printf("*** %08X %08X %04X %02X %08X %02X %08X - %s\n", /* (TEST/DEBUG) */
                l->debug.lock_pc,
                l->debug.lock_thread,
                l->debug.state,
                l->debug.lock_cpu,
                l->debug.unlock_thread,
                l->debug.unlock_cpu,
                l->debug.unlock_pc,
                caller);
#endif

        if (!usld_lock_common_checks(l, caller))
                return;

        if (!(l->debug.state & USLOCK_TAKEN))
                panic("%s:  lock 0x%x hasn't been taken",
                      caller, (integer_t) l);
        if (l->debug.lock_thread != (void *) current_thread())
                panic("%s:  lock 0x%x is owned by thread 0x%x", caller,
                      (integer_t) l, (integer_t) l->debug.lock_thread);

        /*
         *      The usimple_lock is active, so preemption
         *      is disabled and the current cpu should
         *      match the one recorded at lock acquisition time.
         */
        if (l->debug.lock_cpu != cpu_number())
                panic("%s:  current cpu 0x%x isn't acquiring cpu 0x%x",
                      caller, cpu_number(), (integer_t) l->debug.lock_cpu);
}


/*
 *      Per-cpu stack of currently active usimple_locks.
 *      Requires spl protection so that interrupt-level
 *      locks plug-n-play with their thread-context friends.
 */
#define USLOCK_STACK_DEPTH      20
usimple_lock_t  uslock_stack[NCPUS][USLOCK_STACK_DEPTH];
unsigned int    uslock_stack_index[NCPUS];
boolean_t       uslock_stack_enabled = FALSE;


/*
 *      Record a usimple_lock just acquired on
 *      the current processor.
 *
 *      Preemption has been disabled by lock
 *      acquisition, so it's safe to use the cpu number
 *      specified by the caller.
 */
void
usl_stack_push(
        usimple_lock_t  l,
        int             mycpu)
{
        boolean_t       s;

        if (uslock_stack_enabled == FALSE)
                return;

        DISABLE_INTERRUPTS(s);
        assert(uslock_stack_index[mycpu] >= 0);
        assert(uslock_stack_index[mycpu] < USLOCK_STACK_DEPTH);
        if (uslock_stack_index[mycpu] >= USLOCK_STACK_DEPTH) {
                printf("usl_stack_push (cpu 0x%x):  too many locks (%d)",
                       mycpu, uslock_stack_index[mycpu]);
                printf(" disabling stacks\n");
                uslock_stack_enabled = FALSE;
                ENABLE_INTERRUPTS(s);
                return;
        }
        uslock_stack[mycpu][uslock_stack_index[mycpu]] = l;
        uslock_stack_index[mycpu]++;
        ENABLE_INTERRUPTS(s);
}


/*
 *      Eliminate the entry for a usimple_lock
 *      that had been active on the current processor.
 *
 *      Preemption has been disabled by lock
 *      acquisition, and we haven't yet actually
 *      released the hardware lock associated with
 *      this usimple_lock, so it's safe to use the
 *      cpu number supplied by the caller.
 */
void
usl_stack_pop(
        usimple_lock_t  l,
        int             mycpu)
{
        unsigned int    i, index;
        boolean_t       s;

        if (uslock_stack_enabled == FALSE)
                return;

        DISABLE_INTERRUPTS(s);
        assert(uslock_stack_index[mycpu] > 0);
        assert(uslock_stack_index[mycpu] <= USLOCK_STACK_DEPTH);
        if (uslock_stack_index[mycpu] == 0) {
                printf("usl_stack_pop (cpu 0x%x):  not enough locks (%d)",
                       mycpu, uslock_stack_index[mycpu]);
                printf(" disabling stacks\n");
                uslock_stack_enabled = FALSE;
                ENABLE_INTERRUPTS(s);
                return;
        }
        index = --uslock_stack_index[mycpu];
        for (i = 0; i <= index; ++i) {
                if (uslock_stack[mycpu][i] == l) {
                        if (i != index)
                                uslock_stack[mycpu][i] =
                                        uslock_stack[mycpu][index];
                        ENABLE_INTERRUPTS(s);
                        return;
                }
        }
        ENABLE_INTERRUPTS(s);
        panic("usl_stack_pop:  can't find usimple_lock 0x%x", l);
}


/*
 *      Determine whether any usimple_locks are currently held.
 *
 *      Caller's preemption state is uncertain.  If
 *      preemption has been disabled, this check is accurate.
 *      Otherwise, this check is just a guess.  We do the best
 *      we can by disabling scheduler interrupts, so at least
 *      the check is accurate w.r.t. whatever cpu we're running
 *      on while in this routine.
 */
void
usld_lock_none_held()
{
        register int    mycpu;
        boolean_t       s;
        unsigned int    locks_held;
        char            *caller = "usimple_lock_none_held";

        DISABLE_INTERRUPTS(s);
        mp_disable_preemption();
        mycpu = cpu_number();
        locks_held = uslock_stack_index[mycpu];
        mp_enable_preemption();
        ENABLE_INTERRUPTS(s);
        if (locks_held > 0)
                panic("%s:  no locks should be held (0x%x locks held)",
                      caller, (integer_t) locks_held);
}


/*
 *      For very special cases, set traced_lock to point to a
 *      specific lock of interest.  The result is a series of
 *      XPRs showing lock operations on that lock.  The lock_seq
 *      value is used to show the order of those operations.
 */
usimple_lock_t          traced_lock;
unsigned int            lock_seq;

void
usl_trace(
        usimple_lock_t  l,
        int             mycpu,
        pc_t            pc,
        const char *    op_name)
{
        if (traced_lock == l) {
                XPR(XPR_SLOCK,
                    "seq %d, cpu %d, %s @ %x\n",
                    (integer_t) lock_seq, (integer_t) mycpu,
                    (integer_t) op_name, (integer_t) pc, 0);
                lock_seq++;
        }
}



#if     MACH_KDB
#define printf  kdbprintf
void    db_show_all_slocks(void);
void
db_show_all_slocks(void)
{
        unsigned int    i, index;
        int             mycpu = cpu_number();
        usimple_lock_t  l;

        if (uslock_stack_enabled == FALSE) {
                printf("Lock stack not enabled\n");
                return;
        }

#if     0
        if (!mach_slocks_init)
                iprintf("WARNING: simple locks stack may not be accurate\n");
#endif
        assert(uslock_stack_index[mycpu] >= 0);
        assert(uslock_stack_index[mycpu] <= USLOCK_STACK_DEPTH);
        index = uslock_stack_index[mycpu];
        for (i = 0; i < index; ++i) {
                l = uslock_stack[mycpu][i];
                iprintf("%d: ", i);
                db_printsym((vm_offset_t)l, DB_STGY_ANY);
                if (l->debug.lock_pc != INVALID_PC) {
                        printf(" locked by ");
                        db_printsym((int)l->debug.lock_pc, DB_STGY_PROC);
                }
                printf("\n");
        }
}
#endif  /* MACH_KDB */

#endif  /* USLOCK_DEBUG */

/* #endif       USIMPLE_LOCK_CALLS */

/*
 *      Routine:        lock_alloc
 *      Function:
 *              Allocate a lock for external users who cannot
 *              hard-code the structure definition into their
 *              objects.
 *              For now just use kalloc, but a zone is probably
 *              warranted.
 */
lock_t *
lock_alloc(
        boolean_t       can_sleep,
        etap_event_t    event,
        etap_event_t    i_event)
{
        lock_t          *l;

        if ((l = (lock_t *)kalloc(sizeof(lock_t))) != 0)
          lock_init(l, can_sleep, event, i_event);
        return(l);
}

/*
 *      Routine:        lock_free
 *      Function:
 *              Free a lock allocated for external users.
 *              For now just use kfree, but a zone is probably
 *              warranted.
 */
void
lock_free(
        lock_t          *l)
{
        kfree((vm_offset_t)l, sizeof(lock_t));
}

          
/*
 *      Routine:        lock_init
 *      Function:
 *              Initialize a lock; required before use.
 *              Note that clients declare the "struct lock"
 *              variables and then initialize them, rather
 *              than getting a new one from this module.
 */
void
lock_init(
        lock_t          *l,
        boolean_t       can_sleep,
        etap_event_t    event,
        etap_event_t    i_event)
{
        (void) memset((void *) l, 0, sizeof(lock_t));

#if     ETAP_LOCK_TRACE
        etap_event_table_assign(&l->u.event_table_chain, event);
        l->u.s.start_list = SD_ENTRY_NULL;
#endif  /* ETAP_LOCK_TRACE */

        simple_lock_init(&l->interlock, i_event);
        l->want_write = FALSE;
        l->want_upgrade = FALSE;
        l->read_count = 0;
        l->can_sleep = can_sleep;

#if     ETAP_LOCK_ACCUMULATE
        l->cbuff_write = etap_cbuff_reserve(lock_event_table(l));
        if (l->cbuff_write != CBUFF_ENTRY_NULL) {
                l->cbuff_write->event    = event;
                l->cbuff_write->instance = (unsigned long) l;
                l->cbuff_write->kind     = WRITE_LOCK;
        }
        l->cbuff_read = CBUFF_ENTRY_NULL;
#endif  /* ETAP_LOCK_ACCUMULATE */
}


/*
 *      Sleep locks.  These use the same data structure and algorithm
 *      as the spin locks, but the process sleeps while it is waiting
 *      for the lock.  These work on uniprocessor systems.
 */

#define DECREMENTER_TIMEOUT 1000000

void
lock_write(
        register lock_t * l)
{
        register int       i;
        start_data_node_t  entry     = {0};
        boolean_t          lock_miss = FALSE;
        unsigned short     dynamic   = 0;
        unsigned short     trace     = 0;
        etap_time_t        total_time;
        etap_time_t        stop_wait_time;
        pc_t               pc;
#if     MACH_LDEBUG
        int                decrementer;
#endif  /* MACH_LDEBUG */


        ETAP_STAMP(lock_event_table(l), trace, dynamic);
        ETAP_CREATE_ENTRY(entry, trace);
        MON_ASSIGN_PC(entry->start_pc, pc, trace);

        simple_lock(&l->interlock);

        /*
         *  Link the new start_list entry
         */
        ETAP_LINK_ENTRY(l, entry, trace);

#if     MACH_LDEBUG
        decrementer = DECREMENTER_TIMEOUT;
#endif  /* MACH_LDEBUG */

        /*
         *      Try to acquire the want_write bit.
         */
        while (l->want_write) {
                if (!lock_miss) {
                        ETAP_CONTENTION_TIMESTAMP(entry, trace);
                        lock_miss = TRUE;
                }

                i = lock_wait_time[l->can_sleep ? 1 : 0];
                if (i != 0) {
                        simple_unlock(&l->interlock);
#if     MACH_LDEBUG
                        if (!--decrementer)
                                Debugger("timeout - want_write");
#endif  /* MACH_LDEBUG */
                        while (--i != 0 && l->want_write)
                                continue;
                        simple_lock(&l->interlock);
                }

                if (l->can_sleep && l->want_write) {
                        l->waiting = TRUE;
                        ETAP_SET_REASON(current_thread(),
                                        BLOCKED_ON_COMPLEX_LOCK);
                        thread_sleep_simple_lock((event_t) l,
                                        simple_lock_addr(l->interlock), FALSE);
                        simple_lock(&l->interlock);
                }
        }
        l->want_write = TRUE;

        /* Wait for readers (and upgrades) to finish */

#if     MACH_LDEBUG
        decrementer = DECREMENTER_TIMEOUT;
#endif  /* MACH_LDEBUG */
        while ((l->read_count != 0) || l->want_upgrade) {
                if (!lock_miss) {
                        ETAP_CONTENTION_TIMESTAMP(entry,trace);
                        lock_miss = TRUE;
                }

                i = lock_wait_time[l->can_sleep ? 1 : 0];
                if (i != 0) {
                        simple_unlock(&l->interlock);
#if     MACH_LDEBUG
                        if (!--decrementer)
                                Debugger("timeout - wait for readers");
#endif  /* MACH_LDEBUG */
                        while (--i != 0 && (l->read_count != 0 ||
                                            l->want_upgrade))
                                continue;
                        simple_lock(&l->interlock);
                }

                if (l->can_sleep && (l->read_count != 0 || l->want_upgrade)) {
                        l->waiting = TRUE;
                        ETAP_SET_REASON(current_thread(),
                                        BLOCKED_ON_COMPLEX_LOCK);
                        thread_sleep_simple_lock((event_t) l,
                                simple_lock_addr(l->interlock), FALSE);
                        simple_lock(&l->interlock);
                }
        }

        /*
         *  do not collect wait data if either the lock
         *  was free or no wait traces are enabled.
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace)) {
                ETAP_TIMESTAMP(stop_wait_time);
                ETAP_TOTAL_TIME(total_time,
                                stop_wait_time,
                                entry->start_wait_time);
                CUM_WAIT_ACCUMULATE(l->cbuff_write, total_time, dynamic, trace);
                MON_DATA_COLLECT(l,
                                 entry,
                                 total_time,
                                 WRITE_LOCK,
                                 MON_CONTENTION,
                                 trace);
        }

        simple_unlock(&l->interlock);

        /*
         *  Set start hold time if some type of hold tracing is enabled.
         *
         *  Note: if the stop_wait_time was already stamped, use
         *      it as the start_hold_time instead of doing an
         *      expensive bus access.
         *
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace))
                ETAP_COPY_START_HOLD_TIME(entry, stop_wait_time, trace);
        else
                ETAP_DURATION_TIMESTAMP(entry, trace);

}

void
lock_done(
        register lock_t * l)
{
        boolean_t         do_wakeup = FALSE;
        start_data_node_t entry;
        unsigned short    dynamic   = 0;
        unsigned short    trace     = 0;
        etap_time_t       stop_hold_time;
        etap_time_t       total_time;
        unsigned long     lock_kind;
        pc_t              pc;


        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        simple_lock(&l->interlock);

        if (l->read_count != 0) {
                l->read_count--;
                lock_kind = READ_LOCK;
        }
        else    
                if (l->want_upgrade) {
                        l->want_upgrade = FALSE;
                        lock_kind = WRITE_LOCK;
                }
        else {
                l->want_write = FALSE;
                lock_kind = WRITE_LOCK;
        }

        /*
         *      There is no reason to wakeup a waiting thread
         *      if the read-count is non-zero.  Consider:
         *              we must be dropping a read lock
         *              threads are waiting only if one wants a write lock
         *              if there are still readers, they can't proceed
         */

        if (l->waiting && (l->read_count == 0)) {
                l->waiting = FALSE;
                do_wakeup = TRUE;
        }
        /*
         *  Collect hold data if hold tracing is
         *  enabled.
         */

        /*
         *  NOTE: All complex locks whose tracing was on when the
         *  lock was acquired will have an entry in the start_data
         *  list.
         */

        ETAP_UNLINK_ENTRY(l,entry);
        if (ETAP_DURATION_ENABLED(trace) && entry != SD_ENTRY_NULL) {
                ETAP_TIMESTAMP (stop_hold_time);
                ETAP_TOTAL_TIME (total_time,
                                 stop_hold_time,
                                 entry->start_hold_time);

                if (lock_kind & WRITE_LOCK)
                        CUM_HOLD_ACCUMULATE (l->cbuff_write,
                                             total_time,
                                             dynamic,
                                             trace);
                else {
                        CUM_READ_ENTRY_RESERVE(l,l->cbuff_read,trace);
                        CUM_HOLD_ACCUMULATE (l->cbuff_read,
                                             total_time,
                                             dynamic,
                                             trace);
                }
                MON_ASSIGN_PC(entry->end_pc,pc,trace);
                MON_DATA_COLLECT(l,entry,
                                 total_time,
                                 lock_kind,
                                 MON_DURATION,
                                 trace);
        }

        simple_unlock(&l->interlock);

        ETAP_DESTROY_ENTRY(entry);

        if (do_wakeup)
                thread_wakeup((event_t) l);
}

void
lock_read(
        register lock_t * l)
{
        register int        i;
        start_data_node_t   entry     = {0};
        boolean_t           lock_miss = FALSE;
        unsigned short      dynamic   = 0;
        unsigned short      trace     = 0;
        etap_time_t         total_time;
        etap_time_t         stop_wait_time;
        pc_t                pc;
#if     MACH_LDEBUG
        int                decrementer;
#endif  /* MACH_LDEBUG */

        ETAP_STAMP(lock_event_table(l), trace, dynamic);
        ETAP_CREATE_ENTRY(entry, trace);
        MON_ASSIGN_PC(entry->start_pc, pc, trace);

        simple_lock(&l->interlock);

        /*
         *  Link the new start_list entry
         */
        ETAP_LINK_ENTRY(l,entry,trace);

#if     MACH_LDEBUG
        decrementer = DECREMENTER_TIMEOUT;
#endif  /* MACH_LDEBUG */
        while (l->want_write || l->want_upgrade) {
                if (!lock_miss) {
                        ETAP_CONTENTION_TIMESTAMP(entry, trace);
                        lock_miss = TRUE;
                }

                i = lock_wait_time[l->can_sleep ? 1 : 0];

                if (i != 0) {
                        simple_unlock(&l->interlock);
#if     MACH_LDEBUG
                        if (!--decrementer)
                                Debugger("timeout - wait no writers");
#endif  /* MACH_LDEBUG */
                        while (--i != 0 && (l->want_write || l->want_upgrade))
                                continue;
                        simple_lock(&l->interlock);
                }

                if (l->can_sleep && (l->want_write || l->want_upgrade)) {
                        l->waiting = TRUE;
                        thread_sleep_simple_lock((event_t) l,
                                        simple_lock_addr(l->interlock), FALSE);
                        simple_lock(&l->interlock);
                }
        }

        l->read_count++;

        /*
         *  Do not collect wait data if the lock was free
         *  or if no wait traces are enabled.
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace)) {
                ETAP_TIMESTAMP(stop_wait_time);
                ETAP_TOTAL_TIME(total_time,
                                stop_wait_time,
                                entry->start_wait_time);
                CUM_READ_ENTRY_RESERVE(l, l->cbuff_read, trace);
                CUM_WAIT_ACCUMULATE(l->cbuff_read, total_time, dynamic, trace);
                MON_DATA_COLLECT(l,
                                 entry,
                                 total_time,
                                 READ_LOCK,
                                 MON_CONTENTION,
                                 trace);
        }
        simple_unlock(&l->interlock);

        /*
         *  Set start hold time if some type of hold tracing is enabled.
         *
         *  Note: if the stop_wait_time was already stamped, use
         *        it instead of doing an expensive bus access.
         *
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace))
                ETAP_COPY_START_HOLD_TIME(entry, stop_wait_time, trace);
        else
                ETAP_DURATION_TIMESTAMP(entry,trace);
}


/*
 *      Routine:        lock_read_to_write
 *      Function:
 *              Improves a read-only lock to one with
 *              write permission.  If another reader has
 *              already requested an upgrade to a write lock,
 *              no lock is held upon return.
 *
 *              Returns TRUE if the upgrade *failed*.
 */

boolean_t
lock_read_to_write(
        register lock_t * l)
{
        register int        i;
        boolean_t           do_wakeup = FALSE;
        start_data_node_t   entry     = {0};
        boolean_t           lock_miss = FALSE;
        unsigned short      dynamic   = 0;
        unsigned short      trace     = 0;
        etap_time_t         total_time;
        etap_time_t         stop_time;
        pc_t                pc;
#if     MACH_LDEBUG
        int                decrementer;
#endif  /* MACH_LDEBUG */


        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        simple_lock(&l->interlock);

        l->read_count--;        

        /*
         *  Since the read lock is lost whether the write lock
         *  is acquired or not, read hold data is collected here.
         *  This, of course, is assuming some type of hold
         *  tracing is enabled.
         *
         *  Note: trace is set to zero if the entry does not exist.
         */

        ETAP_FIND_ENTRY(l, entry, trace);

        if (ETAP_DURATION_ENABLED(trace)) {
                ETAP_TIMESTAMP(stop_time);
                ETAP_TOTAL_TIME(total_time, stop_time, entry->start_hold_time);
                CUM_HOLD_ACCUMULATE(l->cbuff_read, total_time, dynamic, trace);
                MON_ASSIGN_PC(entry->end_pc, pc, trace);
                MON_DATA_COLLECT(l,
                                 entry,
                                 total_time,
                                 READ_LOCK,
                                 MON_DURATION,
                                 trace);
        }

        if (l->want_upgrade) {
                /*
                 *      Someone else has requested upgrade.
                 *      Since we've released a read lock, wake
                 *      him up.
                 */
                if (l->waiting && (l->read_count == 0)) {
                        l->waiting = FALSE;
                        do_wakeup = TRUE;
                }

                ETAP_UNLINK_ENTRY(l, entry);
                simple_unlock(&l->interlock);
                ETAP_DESTROY_ENTRY(entry);

                if (do_wakeup)
                        thread_wakeup((event_t) l);
                return (TRUE);
        }

        l->want_upgrade = TRUE;

        MON_ASSIGN_PC(entry->start_pc, pc, trace);

#if     MACH_LDEBUG
        decrementer = DECREMENTER_TIMEOUT;
#endif  /* MACH_LDEBUG */
        while (l->read_count != 0) {
                if (!lock_miss) {
                        ETAP_CONTENTION_TIMESTAMP(entry, trace);
                        lock_miss = TRUE;
                }

                i = lock_wait_time[l->can_sleep ? 1 : 0];

                if (i != 0) {
                        simple_unlock(&l->interlock);
#if     MACH_LDEBUG
                        if (!--decrementer)
                                Debugger("timeout - read_count");
#endif  /* MACH_LDEBUG */
                        while (--i != 0 && l->read_count != 0)
                                continue;
                        simple_lock(&l->interlock);
                }

                if (l->can_sleep && l->read_count != 0) {
                        l->waiting = TRUE;
                        thread_sleep_simple_lock((event_t) l,
                                        simple_lock_addr(l->interlock), FALSE);
                        simple_lock(&l->interlock);
                }
        }

        /*
         *  do not collect wait data if the lock was free
         *  or if no wait traces are enabled.
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace)) {
                ETAP_TIMESTAMP (stop_time);
                ETAP_TOTAL_TIME(total_time, stop_time, entry->start_wait_time);
                CUM_WAIT_ACCUMULATE(l->cbuff_write, total_time, dynamic, trace);
                MON_DATA_COLLECT(l,
                                 entry,
                                 total_time,
                                 WRITE_LOCK,
                                 MON_CONTENTION,
                                 trace);
        }

        simple_unlock(&l->interlock);

        /*
         *  Set start hold time if some type of hold tracing is enabled
         *
         *  Note: if the stop_time was already stamped, use
         *        it as the new start_hold_time instead of doing
         *        an expensive VME access.
         *
         */

        if (lock_miss && ETAP_CONTENTION_ENABLED(trace))
                ETAP_COPY_START_HOLD_TIME(entry, stop_time, trace);
        else
                ETAP_DURATION_TIMESTAMP(entry, trace);

        return (FALSE);
}

void
lock_write_to_read(
        register lock_t * l)
{
        boolean_t          do_wakeup = FALSE;
        start_data_node_t  entry   = {0};
        unsigned short     dynamic = 0;
        unsigned short     trace   = 0;
        etap_time_t        stop_hold_time;
        etap_time_t        total_time;
        pc_t               pc;

        ETAP_STAMP(lock_event_table(l), trace,dynamic);

        simple_lock(&l->interlock);

        l->read_count++;
        if (l->want_upgrade)
                l->want_upgrade = FALSE;
        else
                l->want_write = FALSE;

        if (l->waiting) {
                l->waiting = FALSE;
                do_wakeup = TRUE;
        }

        /*
         *  Since we are switching from a write lock to a read lock,
         *  the write lock data is stored and the read lock data
         *  collection begins.
         *
         *  Note: trace is set to zero if the entry does not exist.
         */

        ETAP_FIND_ENTRY(l, entry, trace);

        if (ETAP_DURATION_ENABLED(trace)) {
            ETAP_TIMESTAMP (stop_hold_time);
            ETAP_TOTAL_TIME(total_time, stop_hold_time, entry->start_hold_time);
            CUM_HOLD_ACCUMULATE(l->cbuff_write, total_time, dynamic, trace);
            MON_ASSIGN_PC(entry->end_pc, pc, trace);
            MON_DATA_COLLECT(l,
                             entry,
                             total_time,
                             WRITE_LOCK,
                             MON_DURATION,
                             trace);
        }

        simple_unlock(&l->interlock);

        /*
         *  Set start hold time if some type of hold tracing is enabled
         *
         *  Note: if the stop_hold_time was already stamped, use
         *        it as the new start_hold_time instead of doing
         *        an expensive bus access.
         *
         */

        if (ETAP_DURATION_ENABLED(trace))
                ETAP_COPY_START_HOLD_TIME(entry, stop_hold_time, trace);
        else
                ETAP_DURATION_TIMESTAMP(entry, trace);

        MON_ASSIGN_PC(entry->start_pc, pc, trace);

        if (do_wakeup)
                thread_wakeup((event_t) l);
}


#if     0       /* Unused */
/*
 *      Routine:        lock_try_write
 *      Function:
 *              Tries to get a write lock.
 *
 *              Returns FALSE if the lock is not held on return.
 */

boolean_t
lock_try_write(
        register lock_t * l)
{
        start_data_node_t  entry = {0};
        unsigned short     trace = 0;
        pc_t               pc;

        ETAP_STAMP(lock_event_table(l), trace, trace);
        ETAP_CREATE_ENTRY(entry, trace);

        simple_lock(&l->interlock);

        if (l->want_write || l->want_upgrade || l->read_count) {
                /*
                 *      Can't get lock.
                 */
                simple_unlock(&l->interlock);
                ETAP_DESTROY_ENTRY(entry);
                return(FALSE);
        }

        /*
         *      Have lock.
         */

        l->want_write = TRUE;

        ETAP_LINK_ENTRY(l, entry, trace);

        simple_unlock(&l->interlock);

        MON_ASSIGN_PC(entry->start_pc, pc, trace);
        ETAP_DURATION_TIMESTAMP(entry, trace);

        return(TRUE);
}

/*
 *      Routine:        lock_try_read
 *      Function:
 *              Tries to get a read lock.
 *
 *              Returns FALSE if the lock is not held on return.
 */

boolean_t
lock_try_read(
        register lock_t * l)
{
        start_data_node_t  entry = {0};
        unsigned short     trace = 0;
        pc_t               pc;

        ETAP_STAMP(lock_event_table(l), trace, trace);
        ETAP_CREATE_ENTRY(entry, trace);

        simple_lock(&l->interlock);

        if (l->want_write || l->want_upgrade) {
                simple_unlock(&l->interlock);
                ETAP_DESTROY_ENTRY(entry);
                return(FALSE);
        }

        l->read_count++;

        ETAP_LINK_ENTRY(l, entry, trace);

        simple_unlock(&l->interlock);

        MON_ASSIGN_PC(entry->start_pc, pc, trace);
        ETAP_DURATION_TIMESTAMP(entry, trace);

        return(TRUE);
}
#endif          /* Unused */

#if     MACH_KDB

void    db_show_one_lock(lock_t  *);


void
db_show_one_lock(
        lock_t  *lock)
{
        db_printf("Read_count = 0x%x, %swant_upgrade, %swant_write, ",
                  lock->read_count,
                  lock->want_upgrade ? "" : "!",
                  lock->want_write ? "" : "!");
        db_printf("%swaiting, %scan_sleep\n", 
                  lock->waiting ? "" : "!", lock->can_sleep ? "" : "!");
        db_printf("Interlock:\n");
        db_show_one_simple_lock((db_expr_t)simple_lock_addr(lock->interlock),
                        TRUE, (db_expr_t)0, (char *)0);
}
#endif  /* MACH_KDB */

/*
 * The C portion of the mutex package.  These routines are only invoked
 * if the optimized assembler routines can't do the work.
 */

/*
 *      Routine:        lock_alloc
 *      Function:
 *              Allocate a mutex for external users who cannot
 *              hard-code the structure definition into their
 *              objects.
 *              For now just use kalloc, but a zone is probably
 *              warranted.
 */
mutex_t *
mutex_alloc(
        etap_event_t    event)
{
        mutex_t         *m;

        if ((m = (mutex_t *)kalloc(sizeof(mutex_t))) != 0)
          mutex_init(m, event);
        return(m);
}

/*
 *      Routine:        mutex_free
 *      Function:
 *              Free a mutex allocated for external users.
 *              For now just use kfree, but a zone is probably
 *              warranted.
 */
void
mutex_free(
        mutex_t         *m)
{
        kfree((vm_offset_t)m, sizeof(mutex_t));
}


/*
 * mutex_lock_wait: Invoked if the assembler routine mutex_lock () fails
 * because the mutex is already held by another thread.  Called with the
 * interlock locked and returns with the interlock unlocked.
 */

void
mutex_lock_wait (
        mutex_t         * m)
{
        m->waiters++;
        ETAP_SET_REASON(current_thread(), BLOCKED_ON_MUTEX_LOCK);
        thread_sleep_interlock ((event_t) m, &m->interlock, THREAD_UNINT);
}

/*
 * mutex_unlock_wakeup: Invoked if the assembler routine mutex_unlock ()
 * fails because there are thread(s) waiting for this mutex.  Called and
 * returns with the interlock locked.
 */

void
mutex_unlock_wakeup (
        mutex_t         * m)
{
        assert(m->waiters);
        m->waiters--;
        thread_wakeup_one ((event_t) m);
}

/*
 * mutex_pause: Called by former callers of simple_lock_pause().
 */

void
mutex_pause(void)
{
        int wait_result;

        assert_wait_timeout( 1, THREAD_INTERRUPTIBLE);
        ETAP_SET_REASON(current_thread(), BLOCKED_ON_MUTEX_LOCK);
        wait_result = thread_block((void (*)(void))0);
        if (wait_result != THREAD_TIMED_OUT)
                thread_cancel_timer();
}

#if     MACH_KDB
/*
 * Routines to print out simple_locks and mutexes in a nicely-formatted
 * fashion.
 */

char *simple_lock_labels =      "ENTRY    ILK THREAD   DURATION CALLER";
char *mutex_labels =            "ENTRY    LOCKED WAITERS   THREAD CALLER";

void
db_show_one_simple_lock (
        db_expr_t       addr,
        boolean_t       have_addr,
        db_expr_t       count,
        char            * modif)
{
        simple_lock_t   saddr = (simple_lock_t)addr;

        if (saddr == (simple_lock_t)0 || !have_addr) {
                db_error ("No simple_lock\n");
        }
#if     USLOCK_DEBUG
        else if (saddr->lock_type != USLOCK_TAG)
                db_error ("Not a simple_lock\n");
#endif  /* USLOCK_DEBUG */

        db_printf ("%s\n", simple_lock_labels);
        db_print_simple_lock (saddr);
}

void
db_print_simple_lock (
        simple_lock_t   addr)
{

        db_printf ("%08x %3d", addr, *hw_lock_addr(addr->interlock));
#if     USLOCK_DEBUG
        db_printf (" %08x", addr->debug.lock_thread);
        db_printf (" %08x ", addr->debug.duration[1]);
        db_printsym ((int)addr->debug.lock_pc, DB_STGY_ANY);
#endif  /* USLOCK_DEBUG */
        db_printf ("\n");
}

void
db_show_one_mutex (
        db_expr_t       addr,
        boolean_t       have_addr,
        db_expr_t       count,
        char            * modif)
{
        mutex_t         * maddr = (mutex_t *)addr;

        if (maddr == (mutex_t *)0 || !have_addr)
                db_error ("No mutex\n");
#if     MACH_LDEBUG
        else if (maddr->type != MUTEX_TAG)
                db_error ("Not a mutex\n");
#endif  /* MACH_LDEBUG */

        db_printf ("%s\n", mutex_labels);
        db_print_mutex (maddr);
}

void
db_print_mutex (
        mutex_t         * addr)
{
        db_printf ("%08x %6d %7d",
                   addr, *hw_lock_addr(addr->locked), addr->waiters);
#if     MACH_LDEBUG
        db_printf (" %08x ", addr->thread);
        db_printsym (addr->pc, DB_STGY_ANY);
#endif  /* MACH_LDEBUG */
        db_printf ("\n");
}
#endif  /* MACH_KDB */

#if     MACH_LDEBUG
extern void     meter_simple_lock (
                        simple_lock_t   l);
extern void     meter_simple_unlock (
                        simple_lock_t   l);
extern void     cyctm05_stamp (
                        unsigned long   * start);
extern void     cyctm05_diff (
                        unsigned long   * start,
                        unsigned long   * end,
                        unsigned long   * diff);

#if 0
simple_lock_data_t      loser;
#endif

void
meter_simple_lock(
                simple_lock_t   lp)
{
#if 0
        cyctm05_stamp (lp->duration);
#endif
}

int                     long_simple_lock_crash;
int                     long_simple_lock_time = 0x600;
/*
 *      This is pretty gawd-awful.  XXX
 */
decl_simple_lock_data(extern,kd_tty)

void
meter_simple_unlock(
                simple_lock_t   lp)
{
#if 0
        unsigned long   stime[2], etime[2], delta[2];

        if (lp == &kd_tty)                      /* XXX */
                return;                         /* XXX */

        stime[0] = lp->duration[0];
        stime[1] = lp->duration[1];

        cyctm05_stamp (etime);

        if (etime[1] < stime[1])                /* XXX */
                return;                         /* XXX */

        cyctm05_diff (stime, etime, delta);

        if (delta[1] >= 0x10000)                /* XXX */
                return;                         /* XXX */

        lp->duration[0] = delta[0];
        lp->duration[1] = delta[1];

        if (loser.duration[1] < lp->duration[1])
                loser = *lp;

        assert (!long_simple_lock_crash || delta[1] < long_simple_lock_time);
#endif
}
#endif  /* MACH_LDEBUG */


#if ETAP_LOCK_TRACE

/*
 *  ==============================================================
 *  ETAP hook when initializing a usimple_lock.  May be invoked
 *  from the portable lock package or from an optimized machine-
 *  dependent implementation.
 *  ==============================================================
 */

void
etap_simplelock_init (
        simple_lock_t   l,
        etap_event_t    event)
{
        ETAP_CLEAR_TRACE_DATA(l);
        etap_event_table_assign(&l->u.event_table_chain, event);

#if    ETAP_LOCK_ACCUMULATE
        /* reserve an entry in the cumulative buffer */
        l->cbuff_entry = etap_cbuff_reserve(lock_event_table(l));
        /* initialize the entry if one was returned  */
        if (l->cbuff_entry != CBUFF_ENTRY_NULL) {
                l->cbuff_entry->event    = event;
                l->cbuff_entry->instance = (unsigned long) l;
                l->cbuff_entry->kind     = SPIN_LOCK;
        }
#endif  /* ETAP_LOCK_ACCUMULATE */
}


void
etap_simplelock_unlock(
        simple_lock_t   l)
{
        unsigned short  dynamic = 0;
        unsigned short  trace   = 0;
        etap_time_t     total_time;
        etap_time_t     stop_hold_time;
        pc_t            pc;

        OBTAIN_PC(pc, l);
        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        /*
         *  Calculate & collect hold time data only if
         *  the hold tracing was enabled throughout the
         *  whole operation.  This prevents collection of
         *  bogus data caused by mid-operation trace changes.
         *
         */

        if (ETAP_DURATION_ENABLED(trace) && ETAP_WHOLE_OP(l)) {
                ETAP_TIMESTAMP (stop_hold_time);
                ETAP_TOTAL_TIME(total_time, stop_hold_time,
                                l->u.s.start_hold_time);
                CUM_HOLD_ACCUMULATE(l->cbuff_entry, total_time, dynamic, trace);
                MON_ASSIGN_PC(l->end_pc, pc, trace);
                MON_DATA_COLLECT(l,
                                 l,
                                 total_time,
                                 SPIN_LOCK,
                                 MON_DURATION,
                                 trace);
        }
        ETAP_CLEAR_TRACE_DATA(l);
}

/*  ========================================================================
 *  Since the the simple_lock() routine is machine dependant, it must always
 *  be coded in assembly.  The two hook routines below are used to collect
 *  lock_stat data.
 *  ========================================================================
 */

/*
 *  ROUTINE:    etap_simplelock_miss()
 *
 *  FUNCTION:   This spin lock routine is called upon the first
 *              spin (miss) of the lock.
 *
 *              A timestamp is taken at the beginning of the wait period,
 *              if wait tracing is enabled.
 *
 *
 *  PARAMETERS:
 *              - lock address.
 *              - timestamp address.
 *
 *  RETURNS:    Wait timestamp value.  The timestamp value is later used
 *              by etap_simplelock_hold().
 *
 *  NOTES:      This routine is NOT ALWAYS called.  The lock may be free
 *              (never spinning).  For this reason the pc is collected in
 *              etap_simplelock_hold().
 *
 */
etap_time_t
etap_simplelock_miss (
        simple_lock_t   l)

{
        unsigned short trace     = 0;
        unsigned short dynamic   = 0;
        etap_time_t    start_miss_time;

        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        if (trace & ETAP_CONTENTION)
                ETAP_TIMESTAMP(start_miss_time);

        return(start_miss_time);
}

/*
 *  ROUTINE:    etap_simplelock_hold()
 *
 *  FUNCTION:   This spin lock routine is ALWAYS called once the lock
 *              is acquired.  Here, the contention time is calculated and
 *              the start hold time is stamped.
 *
 *  PARAMETERS:
 *              - lock address.
 *              - PC of the calling function.
 *              - start wait timestamp.
 *
 */

void
etap_simplelock_hold (
        simple_lock_t   l,
        pc_t            pc,
        etap_time_t     start_hold_time)
{
        unsigned short  dynamic = 0;
        unsigned short  trace   = 0;
        etap_time_t     total_time;
        etap_time_t     stop_hold_time;

        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        MON_ASSIGN_PC(l->start_pc, pc, trace);

        /* do not collect wait data if lock was free */
        if (ETAP_TIME_IS_ZERO(start_hold_time) && (trace & ETAP_CONTENTION)) {
                ETAP_TIMESTAMP(stop_hold_time);
                ETAP_TOTAL_TIME(total_time,
                                stop_hold_time,
                                start_hold_time);
                CUM_WAIT_ACCUMULATE(l->cbuff_entry, total_time, dynamic, trace);
                MON_DATA_COLLECT(l,
                                 l,
                                 total_time,
                                 SPIN_LOCK,
                                 MON_CONTENTION,
                                 trace);
                ETAP_COPY_START_HOLD_TIME(&l->u.s, stop_hold_time, trace);
        }
        else
                ETAP_DURATION_TIMESTAMP(&l->u.s, trace);
}

void
etap_mutex_init (
        mutex_t         *l,
        etap_event_t    event)
{
        ETAP_CLEAR_TRACE_DATA(l);
        etap_event_table_assign(&l->u.event_table_chain, event);

#if     ETAP_LOCK_ACCUMULATE
        /* reserve an entry in the cumulative buffer */
        l->cbuff_entry = etap_cbuff_reserve(lock_event_table(l));
        /* initialize the entry if one was returned  */
        if (l->cbuff_entry != CBUFF_ENTRY_NULL) {
                l->cbuff_entry->event    = event;
                l->cbuff_entry->instance = (unsigned long) l;
                l->cbuff_entry->kind     = MUTEX_LOCK;
        }
#endif  /* ETAP_LOCK_ACCUMULATE */
}

etap_time_t
etap_mutex_miss (
        mutex_t    *l)
{
        unsigned short  trace       = 0;
        unsigned short  dynamic     = 0;
        etap_time_t     start_miss_time;

        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        if (trace & ETAP_CONTENTION)
                ETAP_TIMESTAMP(start_miss_time);
        else
                ETAP_TIME_CLEAR(start_miss_time);

        return(start_miss_time);
}

void
etap_mutex_hold (
        mutex_t         *l,
        pc_t            pc,
        etap_time_t     start_hold_time)
{
        unsigned short  dynamic = 0;
        unsigned short  trace   = 0;
        etap_time_t     total_time;
        etap_time_t     stop_hold_time;

        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        MON_ASSIGN_PC(l->start_pc, pc, trace);

        /* do not collect wait data if lock was free */
        if (!ETAP_TIME_IS_ZERO(start_hold_time) && (trace & ETAP_CONTENTION)) {
                ETAP_TIMESTAMP(stop_hold_time);
                ETAP_TOTAL_TIME(total_time,
                                stop_hold_time,
                                start_hold_time);
                CUM_WAIT_ACCUMULATE(l->cbuff_entry, total_time, dynamic, trace);
                MON_DATA_COLLECT(l,
                                 l,
                                 total_time,
                                 MUTEX_LOCK,
                                 MON_CONTENTION,
                                 trace);
                ETAP_COPY_START_HOLD_TIME(&l->u.s, stop_hold_time, trace);
        }
        else
                ETAP_DURATION_TIMESTAMP(&l->u.s, trace);
}

void
etap_mutex_unlock(
        mutex_t         *l)
{
        unsigned short  dynamic = 0;
        unsigned short  trace   = 0;
        etap_time_t     total_time;
        etap_time_t     stop_hold_time;
        pc_t            pc;

        OBTAIN_PC(pc, l);
        ETAP_STAMP(lock_event_table(l), trace, dynamic);

        /*
         *  Calculate & collect hold time data only if
         *  the hold tracing was enabled throughout the
         *  whole operation.  This prevents collection of
         *  bogus data caused by mid-operation trace changes.
         *
         */

        if (ETAP_DURATION_ENABLED(trace) && ETAP_WHOLE_OP(l)) {
                ETAP_TIMESTAMP(stop_hold_time);
                ETAP_TOTAL_TIME(total_time, stop_hold_time,
                                l->u.s.start_hold_time);
                CUM_HOLD_ACCUMULATE(l->cbuff_entry, total_time, dynamic, trace);
                MON_ASSIGN_PC(l->end_pc, pc, trace);
                MON_DATA_COLLECT(l,
                                 l,
                                 total_time,
                                 MUTEX_LOCK,
                                 MON_DURATION,
                                 trace);
        }
        ETAP_CLEAR_TRACE_DATA(l);
}

#endif  /* ETAP_LOCK_TRACE */